Description: We examine the mean pathways, transit timescales, and transformation of waters flowing from the Pacific and the marginal seas through the Indian Ocean (IO) on their way toward the South Atlantic within a high-resolution ocean/sea-ice model. The model fields are analyzed from a Lagrangian perspective where water volumes are tracked as they enter the IO. The IO contributes 12.6 Sv to Agulhas leakage, which within the model is 14.1 ± 2.2 Sv, the rest originates from the South Atlantic. The Indonesian Through-flow constitutes about half of the IO contribution, is surface bound, cools and salinificates as it leaves the basin within 10–30 years. Waters entering the IO south of Australia are at intermediate depths and maintain their temperature-salinity properties as they exit the basin within 15–35 years. Of these waters, the contribution from Tasman leakage is 1.4 Sv. The rest stem from recirculation from the frontal regions of the Southern Ocean. The marginal seas export 1.0 Sv into the Atlantic within 15–40 years, and the waters cool and freshen on-route. However, the model's simulation of waters from the Gulfs of Aden and Oman are too light and hence overly influenced by upper ocean circulations. In the Cape Basin, Agulhas leakage is well mixed. On-route, temperature-salinity transformations occur predominantly in the Arabian Sea and within the greater Agulhas Current region. Overall, the IO exports at least 7.9 Sv from the Pacific to the Atlantic, thereby quantifying the strength of the upper cell of the global conveyor belt.

Description: The Lagrangian analysis of sets of particles advected with the flow fields of ocean models is used to study connectivity, that is, exchange pathways, time scales, and volume transports, between distinct oceanic regions. One important factor influencing the dispersion of fluid particles and, hence, connectivity is the Lagrangian eddy diffusivity, which quantifies the influence of turbulent processes on the rate of particle dispersal. Because of spatial and temporal discretization, turbulence is not fully resolved in modeled velocities, and the concept of eddy diffusivity is used to parameterize the impact of unresolved processes. However, the relations between observation- and model-based Lagrangian eddy diffusivity estimates, as well as eddy parameterizations, are not clear. This study presents an analysis of the spatially variable near-surface lateral eddy diffusivity estimates obtained from Lagrangian trajectories simulated with 5-day mean velocities from an eddy-resolving ocean model (INALT01) for the Agulhas system. INALT01 features diffusive regimes for dynamically different regions, some of which exhibit strong suppression of eddy mixing by mean flow, and it is consistent with the pattern and magnitude of drifter-based eddy diffusivity estimates. Using monthly mean velocities decreases the estimated diffusivities less than eddy kinetic energy, supporting the idea that large and persistent eddy features dominate eddy diffusivities. For a noneddying ocean model (ORCA05), Lagrangian eddy diffusivities are greatly reduced, particularly when the Gent and McWilliams parameterization of mesoscale eddies is employed.

Description: To model tracer spreading in the Ocean, Lagrangian simulations in an offline framework are a practical and efficient alternative to solving the advective-diffusive tracer equations online. Differences in both approaches raise the question whether both methods are comparable. Lagrangian simulations usually use model output averaged in time, and trajectories are not subject to parameterized subgrid diffusion which is included in the advection-diffusion equations of ocean models. Previous studies focused on diffusivity estimates in idealized models but could show that both methods yield similar results as long as the deformations scale dynamics are resolved and a sufficient amount of floats is used. This study compares the spreading of an Eulerian tracer simulated online and a cloud of Lagrangian particles simulated offline with velocities from the same model. We use a global, eddy-resolving ocean model featuring 1/20° horizontal resolution in the Agulhas region around South Africa. Tracer and particles were released at one time step in the Cape Basin and below the mixed layer at a depth of 160 m and integrated for 3 years. Large-scale diagnostics, like mean pathways of floats and tracer, are almost identical and 1D-horizontal distributions show no signigicant differences. Differences in vertical distributions, seen in a reduced vertical spreading and downward displacement of particles, are due to the combined effect of unresolved sub-daily variability of the vertical velocities and the spatial variation of vertical diffusivity. This, in turn, has a small impact on the horizontal spreading behavior. The estimates of eddy diffusivity from particles and tracer yield comparable results of about 4048 m^2/s in the Cape Basin.

Description: July 01 – July 28, 2013 Fortaleza (Brazil) – Walvis Bay (Namibia)

Description: The North Brazil Current (NBC) constitutes a bottleneck for the mean northward return flow of the Atlantic Meridional Overturning Circulation (AMOC) in the tropical South Atlantic. Previous studies suggested a link between interannual to multidecadal NBC and AMOC transport variability and proposed to use NBC observations as an index for the AMOC. Here we use a set of hindcast, sensitivity, and perturbation experiments performed within a hierarchy of ocean general circulation models to show that decadal to multidecadal buoyancy-forced changes in the basin-scale AMOC transport indeed manifest themselves in the NBC. The relation is, however, masked by a strong interannual to decadal wind-driven gyre variability of the NBC. While questioning the NBC transport as a direct index for the AMOC, the results support its potential merit for an AMOC monitoring system, provided that the wind-driven circulation variability is properly accounted for.

Description: Lagrangian connectivity studies with ocean models comprise the analysis of sets of virtual fluid particle trajectories to identify connecting pathways, as well as associated timescales and transports between distinct oceanic regions. They constitute a powerful tool in physical oceanography and a unique means to coherently study seawater pathways associated with the global overturning circulation (GOC). However, there are several open questions related to the (partially unresolved) impact of small-scale flow variability on large-scale Lagrangian connectivity measures. This doctoral dissertation addresses different aspects of the question how high-resolution ocean models can help improving our understanding of the spreading of water masses associated with the global overturning circulation, by: (i) reviewing the theoretical background of Lagrangian connectivity studies with ocean models, thereby highlighting the importance to distinguish Lagrangian analyzes of simulated purely advective volume transport trajectories from Lagrangian modeling approaches to estimate advective-diffusive tracer trajectories; (ii) investigating Lagrangian volume transport pathways and along-track tracer changes of the GOC’s upper limb in the South Atlantic (study 1) and Indian Ocean (study 2) with high-resolution models; and (iii) assessing the performance of near-surface particle dispersal simulations in the extended Agulhas Current system by means of lateral eddy diffusivity estimates (study 3). The major aim of study 1 was to revisit the relative importance of the different sources for the GOC’s upper limb in the South Atlantic, that are, waters entering from the Indian Ocean through the Agulhas Current system versus waters entering from the Pacific through Drake Passage. To do so, backward trajectory calculations from the North Brazil Current have been employed. They revealed a substantially higher volumetric contribution of waters originating at Drake Passage than estimated by previous Lagrangian studies with coarser resolution models. Moreover, the majority of waters with Drake Passage and Agulhas origin experiences substantial water property modifications along their transit through the South Atlantic. A first analysis of temporal changes in the ratio of the two contributions further indicates that an ongoing increase in Agulhas leakage may evoke an increase in the Agulhas contribution to the upper limb of the Atlantic part of the GOC, while the Drake Passage contribution decreases. In study 2 the Agulhas Current contribution to the GOC’s upper limb was further decomposed into its different sources from the Indian Ocean. It constitutes the first coherent Lagrangian analysis of large-scale connectivity associated with the upper limb return flow from the Pacific via the Indian Ocean into the Atlantic. The analysis highlights the dominant contribution of surface-bound Indonesian Throughflow waters that salinify and cool during their transit through the South Atlantic. Additionally, it reports a relative high contribution of intermediate water with origin South of Australia, which largely maintain their properties during transit. Study 3 presents an analysis of near-surface lateral eddy diffusivity estimates obtained from Lagrangian trajectories simulated with a high resolution model for the greater Agulhas system. It further includes the first comparison of eddy diffusivity estimates from simulated Lagrangian trajectories and observed drifter trajectories for that region. Notably, eddy diffusivity estimates from the high-resolution model are consistent with the spatial pattern and magnitude of drifter-based eddy diffusivity estimates — without the need for an additional diffusion parametrization in the trajectory calculation. Overall, these studies contributed to a revised picture of large-scale connectivity associated with the GOC and yielded new insights regarding the suitability of Lagrangian analyzes for assessing large-scale spreading of water masses. In particular, the presented studies highlight the benefits of Lagrangian analyzes to investigate the simulated flow field of ocean models in terms of volume transport pathways and timescales, as well as net water mass transformations. At the same time they indicate that Lagrangian analyzes are of limited use for assessing the pathways and timescales with that upper ocean temperature as well as salinity anomalies may be transmitted through the Indian and South Atlantic oceans, if the virtual fluid particles experience non-negligible along-track water property modifications.

Description: To model tracer spreading in the ocean, Lagrangian simulations in an offline framework are a practical and efficient alternative to solving the advective–diffusive tracer equations online. Differences in both approaches raise the question of whether both methods are comparable. Lagrangian simulations usually use model output averaged in time, and trajectories are not subject to parameterized subgrid diffusion, which is included in the advection–diffusion equations of ocean models. Previous studies focused on diffusivity estimates in idealized models but could show that both methods yield similar results as long as the deformations-scale dynamics are resolved and a sufficient amount of Lagrangian particles is used. This study compares spreading of an Eulerian tracer simulated online and a cloud of Lagrangian particles simulated offline with velocities from the same ocean model. We use a global, eddy-resolving ocean model featuring 1/20° horizontal resolution in the Agulhas region around South Africa. Tracer and particles were released at one time step in the Cape Basin and below the mixed layer and integrated for 3 years. Large-scale diagnostics, like mean pathways of floats and tracer, are almost identical and 1D horizontal distributions show no significant differences. Differences in vertical distributions, seen in a reduced vertical spreading and downward displacement of particles, are due to the combined effect of unresolved subdaily variability of the vertical velocities and the spatial variation of vertical diffusivity. This, in turn, has a small impact on the horizontal spreading behavior. The estimates of eddy diffusivity from particles and tracer yield comparable results of about 4000 m2 s−1 in the Cape Basin.

Description: The northward flow of the upper limb of the Atlantic Meridional Overturning Circulation (AMOC) is fed by waters entering the South Atlantic from the Indian Ocean mainly via the Agulhas Current (AC) system and by waters entering from the Pacific through Drake Passage (DP), commonly referred to as the “warm” and “cold” water routes, respectively. However, there is no final consensus on the relative importance of these two routes for the upper limb's volume transport and thermohaline properties. In this study we revisited the AC and DP contributions by performing Lagrangian analyses between the two source regions and the North Brazil Current (NBC) at 6∘ S in a realistically forced high-resolution (1∕20∘) ocean model. Our results agree with the prevailing conception that the AC contribution is the major source for the upper limb transport of the AMOC in the tropical South Atlantic. However, they also suggest a non-negligible DP contribution of around 40 %, which is substantially higher than estimates from previous Lagrangian studies with coarser-resolution models but now better matches estimates from Lagrangian observations. Moreover, idealized analyses of decadal changes in the DP and AC contributions indicate that the ongoing increase in Agulhas leakage indeed may have induced an increase in the AC contribution to the upper limb of the AMOC in the tropics, while the DP contribution decreased. In terms of thermohaline properties, our study highlights the fact that the AC and DP contributions cannot be unambiguously distinguished by their temperature, as the commonly adopted terminology may imply, but rather by their salinity when entering the South Atlantic. During their transit towards the NBC the bulk of DP waters experiences a net density loss through a net warming, whereas the bulk of AC waters experiences a slight net density gain through a net increase in salinity. Notably, these density changes are nearly completely captured by Lagrangian particle trajectories that reach the surface mixed layer at least once during their transit, which amount to 66 % and 49 % for DP and AC waters, respectively. This implies that more than half of the water masses supplying the upper limb of the AMOC are actually formed within the South Atlantic and do not get their characteristic properties in the Pacific and Indian Oceans.

Description: Ballast water treatment is required for vessels to prevent the introduction of potentially invasive neobiota. Some treatment methods use chemical disinfectants which produce a variety of halogenated compounds as disinfection by-products (DBPs). One of the most abundant DBP from oxidative ballast water treatment is bromoform (CHBr3) where we find an average concentration of 894±560nmolL-1 (226±142μgL-1) in the undiluted ballast water from measurements and literature. Bromoform is a relevant gas for atmospheric chemistry and ozone depletion, especially in the tropics where entrainment into the stratosphere is possible. The spread of DBPs in the tropics over months to years is assessed here for the first time. With Lagrangian trajectories based on the NEMO-ORCA12 model velocity field, we simulate DBP spread in the sea surface and try to quantify the oceanic bromoform concentration and emission to the atmosphere from ballast water discharge at major harbours in the tropical region of Southeast Asia. The exemplary simulations of two important regions, Singapore and the Pearl River Delta, reveal major transport pathways of the DBPs and the anthropogenic bromoform concentrations in the sea surface. Based on our simulations, we expect DBPs to spread into the open ocean, along the coast and also an advection with monsoon-driven currents into the North Pacific and Indian Ocean. Furthermore, anthropogenic bromoform concentrations and emissions are predicted to increase locally around large harbours. In the sea surface around Singapore we estimate an increase in bromoform concentration by 9% compared to recent measurement. In a moderate scenario where 70% of the ballast water is chemically treated bromoform emissions to the atmosphere can locally exceed 1000pmolm-2h-1 and double climatological emissions. In the Pearl River Delta all bromoform is directly outgassed which leads to an additional bromine (Br) input into the atmosphere of 495kmolBr (∼42tCHBr3) a-1. From Singapore ports the additional atmospheric Br input is calculated as 312kmolBr (∼26tCHBr3) a-1. We estimate the global anthropogenic Br input from ballast water into the atmosphere of up to 13Mmola-1. This is 0.1% global Br input from background bromoform emissions and thus probably not relevant for stratospheric ozone depletion.